Predicting prognosis and treatment response of breast cancer patients using expression and cellular localization of N-myristoyltransferase

The instant application is a divisional application of U.S. Ser. No. 18/044,372, which was a 371 of PCT CA 2022/050808, filed May 20, 2022, which claimed the benefit of U.S. Provisional Patent Application 63/190,905, filed May 20, 2021, and entitled “PREDICTING PROGNOSIS AND TREATMENT RESPONSE OF BREAST CANCER PATIENTS USING EXPRESSION AND CELLULAR LOCALIZATION OF N-MYRISTOYLTRANSFERASE”, the entire contents of which are incorporated herein by reference for all purposes.

The human N-myristoyltransferase (NMT) exists in two forms—NMT1 and NMT2. The gene for human NMT1 is located on the long arm of chromosome 17 and the alterative isoforms appear to be splicing variants (Selvakumar P et al.,2007, 46(1):1-36) whereas NMT2 is located on chromosome 10. Previous studies have indicated NMT to be solely cytosolic; however, recently, it has been demonstrated that 10-50% may be associated with a particulate fraction (Boutin J A,1997, 9(1):15-35).

The majority of breast cancers arise from epithelial cells lining the ducts of the breast tissue, and are thus categorized as carcinomas (Sharma, G. N., et al., Journal of Advanced Pharmaceutical Technology & Research, 2010. 1(2): p. 109-126). Central to the cell signalling in most breast carcinomas is estrogen signalling. Estrogen signalling is regulated through the interplay between the two distinct estrogen receptor isoforms (ER α and ER β) and their respective splice variants (Heldring, N., et al., Physiol Rev, 2007. 87(3): p. 905-31). Both ERs are members of the nuclear receptor family of transcription factors, dimerizing upon ligand binding and subsequently localizing to the nucleus to initiate gene transcription (Tamrazi, A., et al., Mol Endocrinol, 2002. 16(12): p. 2706-19). ER activated genes are regulated by regions of DNA collectively known as estrogen response elements (EREs) (Klinge, C. M., Nucleic Acids Research, 2001. 29(14): p. 2905-2919). The primary ER ligand, the steroid hormone estrogen, is a potent morphogen responsible for driving the proliferation of epithelial breast tissues following its binding to EREs, as well as a range of other effects in men and women including those on the cardiovascular, musculoskeletal, immune, and central nervous systems (Gustafsson, J. A., Trends Pharmacol Sci, 2003. 24(9): p. 479-85). Specifically, ER mediated translation produces proteins essential in key processes in breast cancer development, including cell division, survival, and angiogenesis (Osborne, C. K., et al., Clin Cancer Res, 2001. 7(12 Suppl): p. 4338s-4342s). ERs may also participate in “nongenomic” signalling through interaction with proteins in other growth signalling pathways (Losel, R. M., et al., Physiol Rev, 2003. 83(3): p. 965-1016).

Of the different biological estrogen forms, 17β-estradiol (E2) is the most potent version and is the form most frequently involved in breast tissue tumorigenesis (Simpson, E. R., J Steroid Biochem Mol Biol, 2003. 86(3-5): p. 225-30). ER α and ER β are known to have some distinct and divergent functions following the E2 response. This is especially evident at the promoters of important proliferation genes, in which ER α and ER β often have opposing effects (Liu, M. M., et al., J Biol Chem, 2002. 277(27): p. 24353-60). Of the two ER isoforms, ER α overexpression is associated with breast cancer: over half of primary breast cancers exhibit ER α overexpression and approximately 70% of these are sensitive to anti-estrogen therapy (Ali, S. and R. C. Coombes, J Mammary Gland Biol Neoplasia, 2000. 5(3): p. 271-81). Typically, these ER positive breast cancers are treated with either selective estrogen receptor modulators (SERMs), such as tamoxifen, or they are treated with aromatase inhibitors, including anastrozole, exemestane and letrozole (AIs). SERMs generally bind to the ER and act as a competitive inhibitor to block estrogen growth signalling; however, tamoxifen (and other triphenylethylene drugs) does behave as a partial agonist, displaying tissue-selective pharmacology. In fact, evidence suggests that tamoxifen activates the ER, with the subsequent conformational changes of tamoxifen-bound ER resulting in the preferential recruitment of corepressor complexes that lead to gene silencing. Currently, tamoxifen remains the gold standard treatment for primary breast tumors. Due to some of the anti-proliferation effects of ER β signalling, ER β agonists have also been considered in the treatment of some breast cancers (Montanaro, D., et al., J Mol Endocrinol, 2005. 35(2): p. 245-56). The aforementioned treatments are examples of endocrine therapy (also known as hormonal therapy).

ER+ Positive Breast Cancer

Hormone receptor positive breast cancer cells overexpress ER and/or PR, are dependent on the production of endogenous estrogen or progesterone to activate hormone dependent signalling pathways which regulate cellular proliferation rates. The ER has both nuclear (genomic) and non-nuclear (non-genomic) functions and is the major driver of the majority of breast cancers. ER+ breast cancers account for approximately 75% of all diagnosed breast cancer cases (C. K. Osborne and R. Schiff,, vol. 62, pp. 233-247, 2011). As discussed above, ER exist in two isoforms, ERα and ERβ, which belong to the steroid hormone receptor family of nuclear receptors. ERα is the receptor found to be overexpressed in ER+ breast cancer cells and therefore serves as a primary biomarker for ER+ breast cancer prognosis (M. H. Zhang et al.,, vol. 2, no. 1, pp. 41-52, January 2014).

Selective Endocrine Receptor Modulators (SERMs)

As discussed above, there are primarily three classes of agents used to treat ER+ breast tumors: Selective endocrine receptor modulators (SERMs, such as tamoxifen), estrogen synthesis inhibitors (aromatase inhibitors (AIs), such as anastrozole) and selective endocrine receptor down regulators (SERDs, such as fulvestrant) (M. Giuliano et al., The Breast, vol. 20, pp. S42-S49, October 2011). Breast cancer tumors are typically removed by surgery and/or treated with chemotherapy, radiation, and various adjuvant drug therapies such as SERMs. Tamoxifen, which acts as an ER antagonist, competitively inhibiting endogenous estrogen molecules from binding to the ER active site, is considered to be one of the most effective forms of hormonal therapy and is the most commonly prescribed SERM for ER+ breast cancer patients.

Despite the relative success of endocrine therapies in treating breast cancer, de novo and developed resistance to these therapies (endocrine resistance) are still issues of major concern. Almost 50% of breast cancer patients with primary tumors exhibit de novo resistance to first line tamoxifen treatments, with tamoxifen sensitive individuals often acquiring resistance to the drug after an initial positive response. Actual ER expression loss accounts for only a small fraction (10%) of endocrine resistance cases in primary and metastatic tumors (Sighoko, D., et al., Oncologist, 2014. 19(6): p. 592-601). Aberrant PI3K/AKT/mTOR signalling occurs in about 70% of breast cancers, with signalling molecules downstream of the IGF1R receptor also contributing to endocrine resistance, including mutations to the PIK3CA, AKT1, AKT2, PDK1, PTEN and INPP4B genes (Fu, X. et al., Breast, 2013. 22 Suppl 2: p. S12-8).

The PI3K/AKT/mTOR signalling pathway is involved in regulating glucose metabolism, angiogenesis, cell survival, proliferation, and migration, and is often dysregulated in many types of cancer, including breast cancer. The PI3K pathway is triggered by insulin, and growth factors such as EGF, FGF and IGF-1. AKT, which is central to the PI3K pathway, is a serine/threonine protein kinase and proto-oncoprotein (Bellacosa, A., et al., Adv Cancer Res, 2005. 94: p. 29-86). AKT is normally cytoplasmic, but locates to the inner cell membrane by AKT's Plekstrin Homology (PH) domain binding to PIP3, exposing activation sites on AKT for hydroxyl group phosphorylation. Primary phosphorylation of AKT sites for activation are at T308, phosphorylated by phospho-inositide dependent kinase 1 (PDK1), and S473, which is phosphorylated by mTOR complex2 (mTORC2). In ER+ breast cancer cell lines, upon phosphorylation at both T308 and S473, AKT is fully activated and translocates to the cytoplasm, nucleus or other sub-cellular compartments, where it phosphorylates other substrates. A downstream target of AKT is mTOR, which is a kinase that regulates the cellular processes of cell growth, proliferation in response to nutrient/energy availability, signalling stimuli and translation of protein. It has been demonstrated that AKT overexpression leads to decreased NMT activity (Shrivastav, A., et al., J Pathol, 2009. 218(3): p. 391-8). Preliminary studies in our lab have demonstrated that mTOR interacts with and potentially phosphorylates NMT1.

NMT Subcellular Localization

Despite the overlapping targets of NMT1 and NMT2 and their variants, they appear to have different roles in cell apoptosis, during which the myristoylated proteome undergoes drastic changes (Perinpanayagam, M. A., et al., Faseb j, 2013. 27(2): p. 811-2). Ablation of NMT2 has been shown to induce a 2.5× greater rate of apoptosis over NMT1 knockdown in SK-OV-3 ovarian carcinoma cells (Ducker, C. E., et al. Mol Cancer Res, 2005. 3(8): p. 463-76).

Depletion of NMT2 also yielded a shift in BCL family proteins towards a state of apoptosis. These findings support the notion that NMT1 may be the primary NMT involved in driving apoptosis, with NMT2 associating with suites of pro-growth signalling proteins. The same study found that dual depletion of NMT1 and 2 was lethal and that this effect was p53 independent. The line of division between these enzymes' roles may be drawn through their dynamic and differential localization during apoptosis, among other cell states. Perinpanayagam et al demonstrated that both NMT isoforms are cleaved by caspases during apoptosis, in which NMT1 and NMT2 localization changes significantly ((Perinpanayagam, M. A., et al. Faseb j, 2013. 27(2): p. 811-2). NMT1 was shown to be cleaved at Aspartic Acid-72 by either effector caspase 3 or extrinsic caspase 8; NMT2 was shown to be cleaved at Aspartic Acid—25 by effector caspase 3. Caspase-3 is an executioner caspase, which catalyzes the cleavage of many cellular proteins involved in programmed cell death (Nicholson, D. W, Cell Death Differ, 1999. 6(11): p. 1028-42). Following caspase cleavage, which leaves behind a poly-basic domain stretch, a greater population of NMT1 translocated to cytoplasm (55%) from membrane bound whereas NMT2 underwent an even greater shift in localization following caspase cleavage which removed a negatively charged domain, rendering 80% of NMT2 membrane bound as opposed to 62% cytoplasmic prior to caspase cleavage.

Interestingly, serine residues (which are capable of being phosphorylated) within the human NMT isoforms appear to be homologous between different species and between the isoforms themselves. Specifically, serine 47 of NMT1 is similar in relative position to serine 38 of NMT2, with respect to the poly-lysine region in the N-terminus. Additionally, serine 68 which follows the poly-lysine domain of NMT2 is similar in position with serine 73 of NMT1, which has also been identified as phosphorylated in NMT1 following an ultra-deep human phosphoproteome analysis using a human cancer cell line (Sharma, K., et al., Cell Rep, 2014. 8(5): p. 1583-94). The conservation of phosphorylated serines on either side of the poly-lysine region of the NMTs suggests that these residues may play an important role in the regulation of NMT localization within the cell.

According to a first aspect of the invention, there is provided a method of determining prognosis of a hormone positive breast cancer patient comprising:

According to another aspect of the invention, there is provided a method of determining the prognosis of a breast cancer patient comprising:

According to another aspect of the invention, there is provided a method of determining the prognosis of a triple-negative breast cancer patient comprising:

According to another aspect of the invention, there is provided a method of slowing progression or improving outcome of a breast cancer tumor comprising:

According to another aspect of the invention, there is provided a method of identifying a compound capable of inhibiting nuclear translocation of cytoplasmic NMT2 comprising:

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. Although any methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, the preferred methods and materials are now described. All publications mentioned hereunder are incorporated herein by reference.

As discussed herein, high levels of nuclear NMT1 are associated with longer relapse free survival in ERα positive breast cancer patients, subsequently treated with tamoxifen, in univariate analysis. Interestingly, both low levels of cytosolic and nuclear NMT1 correlated to very poor clinical outcomes, as discussed herein.

Specifically, nuclear localization of NMT1, accompanied by low cytoplasmic expression of NMT1, was associated with relapse free recovery following endocrine therapy. The opposite NMT1 localization pattern, high cytoplasmic expression of NMT1, was associated with endocrine therapy resistance and bad prognosis. Thus, we hypothesized that NMT1 is phosphorylated downstream of the PI3K/AKT/mTOR/ER signalling axis by mTOR, and that this phosphorylation event is implicated in NMT1's localization to the nucleus and in good ER+ breast cancer treatment outcome following endocrine therapy. Furthermore, we predicted that the NMT2 isozyme may also be playing an important function in breast cancer signalling, and that this role may also be regulated through phosphorylation.

Furthermore, this study identified and explored potential mTOR or AKT mediated phospho-sites on the NMT1 protein, as well as identified putative phospho-sites on the NMT2 isozyme for future analysis. Datamining using Kinexus Phosphonet and PhosphositePlus revealed that S40 and S47 (NMT1), and S38 and 68 (NMT2) are prime candidates for phosphorylation by mTOR.

Notably, a putative poly-lysine based nuclear localization sequence (NLS) was identified during this study within the N-termini of NMT1 and NMT2. The putative NLS in both proteins is located nearby the residues predicated to be phosphorylated by mTOR. Phosphorylation of NMT may be a possible mechanism in which NMT containing a poly-lysine region is primed for or inhibited from nuclear or ER localization. The presence of S47 phosphorylated NMT1 in the nucleus suggests that this phosphorylation event may be required for nuclear localization of NMT1; the presence of S47 phosphorylated NMT1 following DNA damage suggests that nuclear phospho-NMT1 may be involved in swinging gene transcription towards apoptosis.

Additionally, serine 38 and serines 66, 68 and 70 nearby the poly-lysine region of NMT2 have been shown to be phosphorylated in cancerous tissues (Stuart, S. A., et al., A, Mol Cell Proteomics, 2015. 14(6): p. 1599-615; Zanivan, S., et al., J Proteome Res, 2008. 7(12): p. 5314-26).

It is apparent that most predicted and/or observed phosphorylation events on the NMT isozymes are concentrated around the putative NLS within the N-terminus, regardless of whether they are predicted to be mediated by mTOR. Phosphorylation adjacent to an NLS is a well-known mechanism to regulate importin a mediated translocation of a protein to the inside of the nucleus (Harreman, M. T., et al., J Biol Chem, 2004. 279(20): p. 20613-21). It is plausible that the predicted phosphorylation events may stabilize the structure of the nearby NLS and keep it exposed, as the N-termini of both NMT1 and NMT2 are normally highly disordered.

The S47 residue proved to be a good candidate to study the structural stabilization of the NLS following phosphorylation. Furthermore, phospho-sites tend to be present in the disordered regions of proteins, a pattern that is apparent in NMT1 and NMT2, suggesting that the disordered N-terminal region of NMTs is the primary regulatory region of the enzyme (Landry, C. R., E. D. Levy, and S. W. Michnick, Trends Genet, 2009. 25(5): p. 193-7). This disordered N-terminal region is exposed and highly flexible, making it an easy target for potentially stabilizing phosphorylation. Indeed, the multi-phosphorylation model of NMT1 in which the putative phospho-sites surrounding the NLS were phosphorylated, predicted that the N-terminus of the protein stabilized to form an exposed structure resembling a Helix-Turn-Helix motif. This type of motif is associated with DNA binding and is common to many transcription factors.

MCF7 breast cancer cell lines were established that express various mutant versions of NMT1 fused to a GFP tag. These include variants with either null mutations to the potential phospho-sites (S40A, S47A, and S256A) or phosphorylation mimicking mutations to the sites (S40E. S47E, S256E). We hypothesized that phosphorylation of these sites was involved in shuttling NMT1 to the nucleus or endoplasmic membrane system. Thus, we expected that mutating these sites to alanine phospho-knockouts would result in an NMT1-GFP fusion protein that remained primarily in the cytoplasm. Overall, our prediction was observed, with S40A, S47A and S256A expressing the fusion protein diffusely through the cytoplasm. Inversely, we predicted that at least one of the glutamic acid mutations would result in at least one cell line that exclusively expressed NMT1 in the nucleus; however, results were mixed. Localization that appeared to overlap with the nucleus was observed to a certain degree in S40, S47, and S256 phosphomimics; however, many cells in these populations expressed cytoplasmic NMT1. Unlike the wildtype NMT1-GFP or the alanine mutants, which expressed fusion protein evenly throughout the cytoplasm, S40E and S256E cells that expressed cytoplasmic fusion protein did so in localized areas. These areas often constituted a patch of expression adjacent to the nuclear region, indicating potential localization to the ER. Overall, these findings suggest that phosphorylation of all three sites may be somehow involved in translocation of NMT1 to the nucleus or nuclear membrane, with phosphorylation of S40 and/or S256 involved in translocation of NMT1 to the ER. Within the nucleus, it is possible that NMT1 is playing a role in transcriptional regulation.

The observation of nuclear NMT1, coupled with the identification of a putative DNA interacting NLS, sparked our interest in exploring NMT1's role in the nucleus. We predicted that NMT1 might be interacting with a myristoylated transcriptional co-repressor, BASP1. Immunoprecipitation of BASP1 co-immunoprecipitated NMT1 protein.

Confirmation of the BASP1-NMT1 interaction within the nucleus of MCF7 breast cancer cells led us to investigate a potential interaction of NMT1 with DNA. We showed for the first time through ChIP analysis that NMT1 appears to interact with the P21 and IGF1R growth genes, repression targets of BASP1. Both P21 and IGF1R expression are driving factors in the progression of many cancers, including breast cancer. It is possible that the association of nuclear NMT1 with good breast cancer prognosis is due in part to repression of these and other growth genes.

Although NMT1 and NMT2 are not redundant in function, they share 77% amino acid sequence homology with analogous putatively phosphorylated serine residues.

As discussed herein, NMT2 phosphorylation status is a key element in the progression of ER+ breast cancer cells.

As used herein, “prognosis” refers to for example a “best estimate” of how a cancer will affect a patient, that is, the likely outcome of the cancer. As will be appreciated by those of skill in the art, there are many methods for assigning a prognostic score for a particular patient, that relies on many factors. Accordingly, as used herein, “determining prognosis” refers to the fact that the levels and/or subcellular location of NMT1 and/or NMT2 may represent one prognostic factor in an overall prognosis determination. As such, referring to the prognosis as being “good” or “favorable” indicates that the observed levels and/or subcellular location of NMT1 and/or NMT2 contribute positively to a prognostic score whereas referring to the prognosis as “poor” indicates that a negative contribution is being made to the prognostic score and referring to the prognosis as “worse” indicates that a more negative contribution is being made to the prognostic score. That is, a “worse” prognosis means that the prognostic score is reduced whereas prognosis as “good” or “favorable” means that the prognosis score is increased.

According to an aspect of the invention, there is provided a method of determining prognosis of a hormone positive breast cancer patient comprising:

As discussed herein, cytoplasmic levels and nuclear levels of NMT1 can be determined by a variety of means known in the art. For example, in some embodiments, microscopic analysis of at least one cell from the cell sample may be carried out. In these embodiments, the position of NMT1 may be localized, for example, by antibody binding and subsequent immunofluorescence and/or immunohistochemistry. In this manner, overall levels of NMT1 as well as the cellular localization thereof can be determined. Furthermore, one of skill in the art can easily determine if overall levels of NMT1 are high in either the cytoplasm or nucleus of a given cell for example by comparison with or simply based on knowledge of a control. For example, such a control may be a “positive” control from one or more cells known to have high cytoplasmic or nuclear levels of NMT1 or a “negative” control from one or more cells known to have low cytoplasmic or nuclear levels of NMT1.

As will be known by those of skill in the art, one method for carrying out such a determination is called an H or IHC Score. In methods such as this, slides are scored using standard light microscopy. For example, IHC scores are derived from assessment of both average staining intensity across the two tumor cores (scale 0 to 3) and percentage of positive cells (0 to 100%). These two scores, when multiplied, generate an IHC or H-score of 0 to 300. An “H” score higher than 100 is considered high and less than 100 is considered low for nuclear NMT1, whereas, an “H” score higher than 150 is considered high and lower than 150 is considered low for cytoplasmic NMT1.

In some embodiments, if cytoplasmic levels of NMT1 are determined to be high, for example, having an H score or IHC score of greater than 150, as discussed above, this indicates that the response of the patient to endocrine therapy will be poor, meaning that the patient is at risk for cancer recurrence and death due to breast cancer. Furthermore, a patient with this prognosis would be given a systemic cancer treatment, such as for example chemotherapy, and monitored more frequently for possible recurrence, that is, would be scheduled for more frequent doctor visits and/or examinations than would a patient with a “good” or “favorable” prognosis as understood and accepted by those of skill in the art.

In some embodiments, if nuclear levels of NMT1 are determined to be high, that is, for example, an H score of greater than 100, as discussed above, this indicates that the endocrine therapy response of the patient is likely to be good, and that the patient is at significantly lower risk of recurrence and death due to breast cancer. Accordingly, a patient with this outcome can be administered endocrine therapy and monitored less frequently for possible recurrence.

If both nuclear and cytoplasmic NMT1 levels are low (for example, an H score less than 100 or less than 150 respectively), the endocrine therapy response is worse, with significantly higher risk or rate of recurrence and death due to breast cancer. A patient with this outcome should be assigned a more aggressive systemic therapy than endocrine therapy and monitored much more frequently, as discussed above.

According to another aspect of the invention, there is provided a method of determining the prognosis of a breast cancer patient comprising:

For example, a poor prognosis means that the patient is at risk of recurrence or death due to breast cancer within the first ten years of diagnosis and should be given systemic treatment (chemotherapy) and monitored for recurrence more frequently, as discussed herein.

For a patient with a good prognosis, wherein there is no nuclear NMT2, the patient may be put on endocrine therapy (in the case of hormone-positive cancers) alone for the first ten years and then just monitored for the next ten years.

According to another aspect of the invention, there is provided a method of determining the prognosis of a triple-negative breast cancer patient comprising:

If the prognosis is poor, most of the breast cancer patients will die within the first ten years after diagnosis, usually within the first 4 years of diagnosis. These patients should be placed on increased surveillance.

As will be appreciated by one of skill in the art, if there is no nuclear localization of NMT2, the prognosis is better or favorable, as discussed herein. Endocrine therapy will be the adjuvant therapy to all hormone receptor breast cancer irrespective of the NMT2 status. Positive nuclear NMT2 staining will be suggestive of the poor prognosis and recurrence wherein additional combination therapy along with endocrine therapy may benefit the patient.

As discussed herein, after ten years, the nuclear NMT2 doesn't matter much but high cytoplasmic NMT2 still relates to a bad outcome. Specifically, the survival outcome is poor for very high NMT2 in the cytoplasm. This remains true as an intermediate predictor for high NMT2 in the cytoplasm is a poor predictor.

As will be appreciated by one of skill in the art, while in some embodiments, at least one cell of the cell sample is examined, it is preferable that a statistically significant number of cells are analyzed and the results analyzed, for example, averaged. In some embodiments, at least about one hundred cells from the cell sample are examined for determining NMT1 and/or NMT2 levels and/or cellular location or subcellular location as discussed herein.

According to another aspect of the invention, there is provided a method of slowing progression of a breast cancer tumor or improving outcome of a breast cancer treatment comprising:

In some embodiments, the NMT inhibitor is a compound that inhibits both NMT1 and NMT2 activity.

In some embodiments, the NMT inhibitor is an NMT2 inhibitor, that is, specific for inhibition of NMT2 activity.